【作者】 段军棋；

【导师】 何子述；

【作者基本信息】 电子科技大学 ， 信号与信息处理， 2009， 博士

【摘要】 多输入多输出(MIMO)雷达概念先后由麻省理工学院林肯实验室、贝尔实验室和新泽西技术研究所等单位提出,近年来得到了广泛研究,成为雷达领域理论和实验研究的热点。通常MIMO雷达的各个发射天线(或子阵、或阵元)分别发送相互正交的信号,在空间形成低增益宽波束；各个接收天线(或子阵,或阵元)独立地接收信号然后通过信号处理实现信号的合成和积累,称为正交波形MIMO雷达(以下简称正交MIMO雷达)。正交MIMO雷达在目标检测性能、角度测量能力、动态范围和主瓣低截获概率特性等方面优于传统雷达。正交MIMO雷达的波形直接影响雷达的最终性能,只有具有良好特性的波形才能充分发挥MIMO雷达的探测潜力,因此正交波形设计成为正交MIMO雷达系统的重要研究课题。近年来已有一些对正交MIMO雷达波形设计的研究,但现有研究结果在使用中受到不同程度的限制。本文针对正交MIMO雷达波形设计及相关的信号处理进行了深入研究,主要工作和贡献如下：1、正交多频信号设计及处理。研究了通用的正交多频信号形式,列举了几种可能的类型,指出常规OFDM-LFM和正交多相编码均为其特例。提出了针对多频信号的多普勒模糊分辨技术、多普勒积累方法和高分辨多普勒处理算法,并对它们的性能进行了理论或仿真分析。引入多载波相位编码(MCPC)信号及其调幅形式到MIMO雷达中,并给出了一种快速的脉压方法。2、正交噪声信号设计。提出了改善雷达探测性能的噪声信号产生和优化方法。特别地,引入了谱成形技术优化旁瓣和提出了一种控制信号峰值因子的非线性映射,通过调节映射函数的参数能够得到不同的峰值因子。同时又借助于噪声信号本质上具有的良好互相关特性,设计结果的相关性能较现有的MIMO雷达信号如正交离散频率编码信号(DFCW)、正交多相编码信号等得到显著提高。3、正交混沌信号设计。研究了混沌系统参数和初值与雷达探测性能的关系。给出了系统参数优化准则和初值选择的方法,改变模拟信号带宽的两种方法并对滤波处理进行了研究。在具体设计时,首先选择适合雷达基本探测要求的混沌系统及其参数产生信号胚,根据探测要求设计信号带宽,然后通过信号处理优化方法,产生适合正交MIMO雷达探测的信号集。4、步进时间间隔脉冲串(STIPT)波形设计及其信号处理。提出了一种特殊的时间编码信号及其对应的信号处理算法(MCA)。给出了相干MCA处理和双极性非相干MCA处理两种实现结构。相干MCA利用FFT实现快速处理,能够同时无模糊地进行目标距离和多普勒测量；双极性非相干MCA处理运算简单,能够用于不需要多普勒信息、常规非相干雷达或简易雷达系统中。5、基于STIPT信号的正交编码波形设计。提出了三种正交STIPT编码方法,即正交频分编码、正交参差间隔编码和正交随机间隔编码。针对参差间隔编码方法,给出了设计的基本原则并提供了一个设计示例,能够得到理想的距离旁瓣特性。针对正交随机间隔编码模式,提供了一个仿真实验,结果表明通过该方法能够设计较大数量具有优良特性的正交波形,自相关峰值旁瓣电平和互相关峰值电平优于-30dB。上面研究的正交MIMO雷达波形设计及其相关的信号处理方法可直接或经过少量修改应用于双基地、多基地雷达系统中。部分波形的单信号版本,如优化的噪声信号、混沌信号和STIPT信号等,可作为常规雷达的信号,以改进现有雷达的探测性能。更多还原

【Abstract】 The term of multiple-input multiple-output (MIMO) radar has recently been introduced by MIT Lincoln Laboratory, Bell Laboratory and New Jersey Institute of Technology. MIMO radar has been extensively studied in recent years and has become a hot topic of theoretical researches and experiments in the field of radar.In the MIMO radar, all transmitting elements generally send different signals, form a low-gain wide-beam in spatial domain; and all receive elements receive echoes independently and then perform synthetic processing and integration. MIMO radar operated with orthogonal waveforms is called orthogonal waveform MIMO radar.Orthogonal waveform MIMO radar has some advantages on target detection, angle measurement capability and low probability of intercept (LPI) compared with the traditional radar. The performance of orthogonal MIMO radars is partly decided by characteristics of waveforms. Hence, orthogonal waveform design becomes a significant research subject and only excellent waveforms can exploit the potential of MIMO radars. In recent years, some orthogonal waveforms for MIMO radars have been derived. These results, whereas to some extent, have some limits on performance.In this dissertation, orthogonal waveforms and corresponding signal processing techniques for MIMO radar are explored. The main contributions are summarized as follows.1 Orthogonal multi-frequency signals design and processing. A general orthogonal multi-frequency signal form is researched, several possible types are listed. Conventional OFDM-LFM and orthogonal polyphase coding are its special cases. Doppler ambiguous resolusion, Doppler integration and high resolution Doppler processing approaches and their performance analyses are presented for multi-frequency signal. In addition, multi-carrier phase coded-amplitude modulation signal is suggested for MIMO radar and corresponding fast pulse compression approach is also presented.2 Orthogonal noise waveforms design. Some generation and optimization methods for noise are presented. In particular, spectrum shaping-based sidelobe mitigation and nonlinear bending function mapping-based crest factor reduction techniques are proposed. Due to the noise nature, optimized noise is suitable for orthogonal waveforms generation. The design results are much better than orthogonal discrete frequency coding waveform and orthogonal polyphase coding.3 Orthogonal chaotic signals design. The relations between chaotic signal and radar performance are analyzed by the help of the sensitivity to initial conditions and system parameters. Specially, parameters’optimation, selection of initial values and design of band width, and pre-filtering are researched. Chaotic radar signals are generated through designing parameters of chaos system for good signal embryos and band width of signals, and utilizing a set of optimizing algorithms.4 Stepped time intervals pulse train (STIPT) waveform design and corresponding processing. The range sidelobes are eliminated through modified correlation algorithm (MCA). Coherent MCA and Manchester code-based non-coherent MCA architectures are presented. Coherent MCA performs fast processing by using FFT and can obtain unambiguous range and Doppler measurements. The non-coherent MCA can be used in simple or conventional noncoherent radar systems.5 Orthogonal STIPT-based waveforms design. Three orthogonal STIPT-based coding methods, frequency divided STIPT coding, stagger interval STIPT coding and random interval STIPT coding, are proposed for orthogonal MIMO radar waveform generation. The design principle and an example are presented to obtain a perfect range sidelobe performance from stagger interval STIPT coding. A simulation is presented to achieve a large number of orthogonal waveforms from random interval STIPT coding. Both the peak of autocorrelation sidelobe level and the peak of cross-correlation level of these resulted waveforms are superior to -30dB.The above proposed orthogonal waveforms can be used in MIMO radar systems; also can be used to bistatic, multistatic or netted radar systems straightly or through minor modifications. The single waveform versions of some orthogonal signals can be used in the conventional monostatic radar to improve the radar performance.更多还原